Quantum Chemistry Calculation Aided Design of Chiral Ionic

Quantum chemistry calculation aided design of chiral ionic liquid-based extraction system for amlodipine separation

DING, Qi <http://orcid.org/0000-0003-0994-0210>, CUI, Xing, XU, Guo-Hua, HE, Chao-Hong and WU, Kejun

Available from Sheffield Hallam University Research Archive (SHURA) at:

http://shura.shu.ac.uk/22335/

This document is the author deposited version. You are advised to consult the publisher's version if you wish to cite from it.

Published version

DING, Qi, CUI, Xing, XU, Guo-Hua, HE, Chao-Hong and WU, Kejun (2018). Quantum chemistry calculation aided design of chiral ionic liquid-based extraction system for amlodipine separation. AIChE Journal, 64 (11), 4080-4088.

Copyright and re-use policy

See http://shura.shu.ac.uk/information.html

Sheffield Hallam University Research Archivehttp://shura.shu.ac.uk

http://shura.shu.ac.uk/

http://shura.shu.ac.uk/information.html

Quantum Chemistry Calculation Aided Design of Chiral Ionic

Liquid-Based Extraction System for Amlodipine Separation

Qi Dinga, Xing Cui

a, Guo-Hua Xu

a, Chao-Hong He*

a and Ke-Jun Wu*

b,c

aZhejiang Provincial Key Laboratory of Advanced Chemical Engineering Manufacture

Technology, College of Chemical and Biological Engineering, Zhejiang University,

Hangzhou 310027, China

bDepartment of Engineering and Mathematics, Faculty of Arts, Computing, Engineering and

Science, Sheffield Hallam University, City Campus, Howard Street, Sheffield S1 1WB, UK

cMaterials and Engineering Research Institute, Sheffield Hallam University, Howard Street,

Sheffield S1 1WB, UK

Abstract: Amlodipine is a widely used medication in treating hypertension, which is also

known as a chiral compound. So far efforts have been made to obtain optically pure

(S)-amlodipine because (R)-amlodipine has poor efficacy and is related to undesirable side

effects. However, the available separation methods for amlodipine are still unsatisfactory.

Recently, chiral separation has become a promising application of chiral ionic liquids (CILs),

because the structural designability enables them adjustable separation efficiency for specific

tasks. In this work, a high-efficient CIL-based liquid-liquid extraction system was developed

for racemic amlodipine separation with the assistance of quantum chemistry calculations.

Enantioselectivity up to 1.35 achieved by the novel system at 298.15 K is significantly higher

than other available extraction systems. Moreover, the recycling of CIL can be easily realized

by backward extraction of amlodipine, which is important for the industrial application of

CILs.

Keywords: ionic liquids, amlodipine, chiral separation, extraction, computational chemistry

This article is protected by copyright. All rights reserved.

Acc

epte

d A

rticl

e

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1002/aic.16372 © 2018 American Institute of Chemical Engineers (AIChE) Received: Mar 03, 2018; Revised: Jul 15, 2018; Accepted: Jul 30, 2018

Separations: Materials, Devices and ProcessesAIChE Journal DOI 10.1002/aic.16372

Amlodipine is a third-generation dihydropyridine calcium channel antagonist, which has

been widely used for the treatment of cardiovascular diseases such as hypertension and

angina pectoris in clinical trials.1 Known as a chiral compound (Figure 1), its two

enantiomers exhibit different biological and pharmacological responses in internal

environment. According to the present research, the hypotensive activity mainly arises from

(S)-amlodipine, which can inhibit the calcium influx across cell membranes.2 (R)-amlodipine

has very poor potency, while it may release nitric oxide into the peripheral blood vessels,

further resulting in peripheral edema and some other undesirable side effects.3 It’s widely

accepted that to isolate (S)-amlodipine from its racemate and administer amlodipine in the

single enantiomeric form is beneficial for both safety and efficacy of the medication.

The separation of chiral compounds is of substantial significance especially in

pharmaceutical industry, which is nevertheless a challenging task due to the fact that

enantiomers possess identical physical and chemical properties in achiral environment.4, 5

Till

now, numerous attempts have been made to develop enantioselective separation process to

obtain optically pure (S)-amlodipine with emphasis on diverse chromatographic techniques,6,

7 crystallization

8, 9 and extraction.

10, 11 While the chromatographic techniques are only

applicable in laboratory scale considering the low capacity and large consumption of

solvents.12

Preferential crystallization featuring the formation of diastereomeric salts using

tartaric acid derivatives is the most popular technique in industrial separations. But the

disadvantage comes from the extensive use of hazardous organic solvents like DMSO, DMF

and DCM, and it just gives rise to the safety and environmental concerns of the industrial


Acc

epte

d A

rticl

e

separation process. Furthermore, involved in bulk solid-handling operations, the

crystallization separation processes are generally tedious and time-consuming.13

Liquid-liquid extraction is seen as an alternative technology to overcome the deficiencies of

chromatography and crystallization, while the currently available extraction systems

comprising conventional chiral extractants are still subjected to poor efficiency and all

require to be handled at quite frigid temperature, which means intensive energy consumption.

Without doubt, there is a great demand to develop high-efficient and environmentally benign

separation process for amlodipine racemate.

Chiral ionic liquids (CILs) are a subclass of ionic liquids, in which the cation, anion or

seldom both may be chiral. CILs are of importance due to the unique chiral recognition

capability they exhibit,14, 15

besides other glaring merits inherited from ionic liquids such as

low volatility, good chemical stability and excellent solubility.16

Moreover, CILs are proposed

as designable chiral media, because their properties can be finely tuned by modifying the type

and composition of the ion pairs for different tasks to increase the separation efficiency.17

Owing to the favorable advantages, CILs are widely used in diverse separation processes,18-20

especially their application in enantioselective extraction became an important issue in

research in recent years. A pioneering work to separate phenylalanine racemate through chiral

ligand-exchange liquid-liquid extraction using Cu2+

modified amino acid ionic liquid (AAIL)

done by Tang et al achieved an excellent enantiomeric excess value reaching up to 50.6% in

single-stage extraction.21

In other relevant research, the separation efficiency was further

improved by introducing tropine-based cation into the AAIL.22

Despite these results show

that CIL-based extraction is a very promising chiral separation technology, the difficulty in


Acc

epte

d A

rticl

e

further recovery of the products from the metal ligand still remains to be a problem.21

More

importantly, metal contamination on products especially pharmaceuticals may become a

critical hurdle to limit the industrial application of CILs. An effective way to overcome the

defects of chiral ligand-exchange extraction is to exploit the chiral recognition capabilities of

CILs directly without the use of coordinated metals,23

which meanwhile sets higher

requirement for the configuration and chemistry of CILs to match specific chiral compound

and has rarely been reported. Considering in some discussions earlier, quantum chemistry

calculation has been successfully used to investigate the structure-function relationship of

CILs and predict their performance in extraction separation of phenolic homologs,24

we are

inspired to expect that this approach may also provide guidance to the selection of CIL for

enantioselective extraction. Despite other computational methods such as conductor-like

screening model (COSMO) calculation is available to predict behaviors of ionic liquids in

complex systems by considering the charge distribution on molecular surfaces,25, 26

it’s not

applicable to discriminate optical isomers in this work as the difference in charge

distributions between molecules only differing in conformations is usually very slight.27

With the aim to create a high-performance and green separation process for amlodipine,

we report an experimental research combined with quantum chemistry calculations to

develop a CIL-based liquid-liquid extraction system in this work. AAILs derived from

natural amino acids were chosen for investigation as they were low-cost and

biodegradability.21, 28

A pre-screening of AAILs was carried out by quantum chemistry

calculation, and the predictive capability of the theoretical calculations was further validated

by experimental data. The impacts of organic solvent, solution pH, concentration of the


Acc

epte

d A

rticl

e

substrates and extraction temperature on the separation efficiency, along with the recycling of

AAIL were systematically investigated. It is prospected the approach proposed in the present

work may extend the application of CILs in chiral separations.

Methods

Quantum chemistry calculation details

Quantum chemistry calculations in this work were performed with Gaussian 16

software.29

The hybrid Becke 3-Lee-Yang-Parr (B3LYP) exchange-correlation functional30

and the density functional theory (DFT)31

with 6-31+G(d,p) basis set were employed. And

dispersion-corrected32

DFT at B3LYP-D3/6-311+G(d,p) level of theory is also applied to

determine the reasonable geometries. No constraint was applied to the geometry

optimizations and all optimized geometries were verified as local minimums on the potential

energy surface by vibrational frequency calculations. The geometries of amlodipine

enantiomers, anions and cations of AAILs were produced by optimization and modification

based on the previously published configurations.33-36

The optimization of ion pairs was done

by placing the anions at different positions around the cations to construct different initial

conformations and then optimizing them respectively. Conformation with the lowest energy

was taken as global minimum for the AAIL. The initial conformations of amlodipine-AAIL

complexes were constructed based on the molecular electrostatic potentials of the

enantiomers and the ion pairs. Subsequently, the global minimums of all complexes were

explored by geometry optimizations on these initial conformations. Interaction energies

between amlodipine enantiomers and AAILs were calculated and corrected by basis set


Acc

epte

d A

rticl

e

superposition errors (BSSE) obtained by the counterpoise procedure method37

and zero-point

energies (ZPE) obtained within the harmonic approximation. To get further insight into the

interactions, atoms in molecules (AIM) analysis was also performed for the optimized

amlodipine-AAIL complexes by AIMALL program.38

Chemicals and analytical method

Racemic amlodipine was purchased from Dalian Meilun Biotech Co. Ltd (Liaoning,

China). 1-butyl-3-methylimidazolium L-glutamate ([Bmim][Glu], 97%),

1-ethyl-3-methylimidazolium L-glutamate ([Emim][Glu], 97%), 1-butyl

-3-methylimidazolium L-serinate ([Bmim][Ser], 97%), 1-ethyl-3-methylimidazolium

L-serinate ([Emim][Ser], 97%), 1-butyl-3-methylimidazolium L-phenylalanate ([Bmim][Phe],

97%) and 1-ethyl-3-methylimidazolium L-phenylalanate ([Emim][Phe], 97%) were

purchased from Chengjie Chemical Co. Ltd (Shanghai, China). n-decanol (98%) and

n-hexanol (98%) were purchased from Aladdin Reagent Co. Ltd (Shanghai, China).

1,2-dichloroethane (AR, 99.0%), dichloromethane (AR, 99.5%), n-octanol (AR, 99.0%),

sodium acetate (NaAc, AR, 99%) and acetate (HAc, AR, 99.5%) were purchased from

Sinopharm Chemical Reagent Co. Ltd (China). NaAc/HAc buffer solutions in different pH

ranges were prepared by mixing of sodium acetate and acetate solutions (both 0.3 mol/L) in

proper proportions, and the pH measurements were performed with a pHS-3C digital

pH-meter (LeiCi, Shanghai, China).

Concentration of amlodipine enantiomers in organic phase was detected by

high-performance liquid chromatography (HPLC, Agilent 1260 Infinity LC system). An

EnantioPak SCDP column (5 μm, 4.6 × 250 mm2) was used with the mobile phase made up


Acc

epte

d A

rticl

e

of n-hexane (containing 0.1% trifluoroacetic acid) and ethanol (85/15, v/v). The flow rate of

the mobile phase was set to be 1 mL/min, and the detection on amlodipine was performed by

defined wavelength at 237 nm.

Extraction and backward extraction process

In this work, all AAILs were diluted before use, as the pure AAILs were highly viscous

near room temperature. The aqueous phase was prepared by dissolving a known amount of

AAIL into NaAc/HAc buffer solution, and the organic phase was prepared by dissolving

racemic amlodipine into halohydrocarbon or aliphatic alcohol. The aqueous phase and the

organic phase at the same volume were put together into a shake flask, shaken isothermally at

220 rpm for 3 hours when the extraction equilibrium has already been achieved as shown in

Supplementary Figure 1 (take n-decanol-buffer/[Bmim][Glu] system for example), and

settled for 30 minutes. The organic phase was then taken out of the flask without disturbing

the aqueous phase and detected by HPLC to determine the concentration of amlodipine. Due

to all organic solvents used in this work were almost immiscible with water, the concentration

of amlodipine in the aqueous phase could be determined by mass balance method.39-41

All

experiments were repeated at least three times and the errors in distribution coefficients were

less than 5%. The distribution coefficient and enantioselectivity were calculated by the

following equations,

Sa

SoS

C

CD (1)

Ra

RoR

C

CD (2)


Acc

epte

d A

rticl

e

R

S

D

D (3)

where CSo and CRo, CSa and CRa refer to the concentration of (S)-amlodipine and

(R)-amlodipine in the organic phase, concentration of (S)-amlodipine and (R)-amlodipine in

the aqueous phase, respectively. DS and DR refer to the distribution coefficients of

(S)-amlodipine and (R)-amlodipine, and α refers to the enantioselectivity.

To assess the mutual solubility between AAIL and organic solvent during the extraction

process, blank extraction experiments were conducted. All experimental conditions were kept

the same as mentioned above, except that amlodipine racemate was not added to exclude the

interference for detection. After phase equilibrium had been reached, the content of AAIL in

the organic solvent was determined on a 752PC UV–vis spectrophotometer (Shanghai

spectrum, China). The results show that AAIL dissolved in the organic solvent is negligible.

Take n-decanol-buffer (pH=5.5)/[Bmim][Glu] biphasic system for example, the mole fraction

of AAIL in the organic solvent is only nearly 0.01% (Supplementary Figure 2).

After the extraction process, the AAIL containing aqueous phase loaded with amlodipine

was separated from the organic phase and used for backward extraction. The aqueous phase

was mixed with n-decanol (1:1, v/v) and shaken vigorously under 298.15 K for 2 hours to

reach phase equilibrium, then settled for half an hour. The aqueous phase and the organic

phase were separated from each other, and the aqueous phase was reused in the extraction

process. The schematic representation of the experimental procedure is shown in Figure 2.


Acc

epte

d A

rticl

e

Results and Discussion

Hunting AAILs for the separation of amlodipine by quantum chemistry calculations

In this work, the candidate AAILs are generated by combining the imidazolium cations

Emim+, Bmim

+ with amino-acid derived anions Glu

-, Phe

- and Ser

-. At the beginning,

[Bmim][Glu] will be proposed as a representative to illustrate the chiral recognition

mechanisms of AAILs. The optimized geometry of the complex formed by [Bmim][Glu] and

(R)-amlodipine at B3LYP/6-31+G(d,p) level of theory is displayed in Figure 3a. There are

three explicit hydrogen bonds represented by dash lines formed between them, and both the

cation and the anion of the AAIL are involved in these hydrogen bonding interactions. The

oxygen atoms contained in the γ-carboxyl of Glu- are labeled as O66 and O67 respectively, and

they both act as hydrogen bond acceptors for (R)-amlodipine. The corresponding hydrogen

donor for O66 is the hydrogen atom contained in the amino group of (R)-amlodipine, and that

for O67 is the hydrogen atom contained in the methyl, which is connected to the

dihydropyridine ring of (R)-amlodipine. Besides, a third hydrogen bond is found between the

hydrogen atom on the butyl group of Bmim+ and the carbonyl oxygen atom of

(R)-amlodipine, which is labeled as O31. According to the results of AIM analysis, the

electron densities (ρc) on bond critical points (BCPs) of N44-H46…O66, C84-H85…O31 and

C19-H22…O67 are 0.0165, 0.0113 and 0.0058 a.u. respectively, and the corresponding

Laplacian values (▽2ρc) are 0.0653, 0.0423 and 0.0283 a.u.. All these values are within the

typical range of hydrogen bond (0.002-0.035 a.u. for ρc, and 0.024-0.139 a.u. for ▽2ρc).

42


Acc

epte

d A

rticl

e

In the optimized geometry of (S)-amlodipine-[Bmim][Glu] (Figure 3b), interaction

between the γ-carboxyl of Glu- and the amino group of (S)-amlodipine, and that between the

side chain of Bmim+ and the carbonyl oxygen atom O31 of the enantiomer can still be found.

Nevertheless, due to the steric effect between the chlorphenyl group of (S)-amlodipine and

the butyl group of Bmim+, the orientation between the side chains of Bmim

+ and O31 is

different from that in (R)-amlodipine-[Bmim][Glu]. As is shown, the hydrogen atom to

interact with O31 is located in the methyl rather than the butyl of Bmim+. Moreover, there’s

no hydrogen bond between the γ-carboxyl of Glu- and the hydrogen atoms in the methyl of

(S)-amlodipine as in (R)-amlodipine-[Bmim][Glu], because they move further from each

other also by influence of the steric effect. The length of C80-H82…O31 in

(S)-amlodipine-[Bmim][Glu] is 2.43 Å, and it is relatively longer than C84-H85…O31 in

(R)-amlodipine-[Bmim][Glu] by 0.12 Å, indicating the hydrogen bonding interaction

between the side chain of Bmim+ and the carbonyl oxygen atom O31 in

(R)-amlodipine-[Bmim][Glu] should be stronger than in (S)-amlodipine-[Bmim][Glu], since a

shorter length is usually more favorable to forming a more stable hydrogen bond. The

inference is further confirmed by the results of AIM analysis, because the values of ρc and ▽

2ρc for C80-H82…O31 are 0.0092 and 0.0357 a.u. respectively, and both of them are smaller

than those of C84-H85…O31. The length of N44-H46…O67 in (S)-amlodipine-[Bmim][Glu] is

also longer than N44-H46…O66 in (R)-amlodipine-[Bmim][Glu], and further analysis of AIM

reveals N44-H46…O66 possesses higher stability as well, with slightly larger ρc and ▽2ρc than

N44-H46…O67 (0.0162 a.u. for ρc and 0.0635 a.u. for ▽2ρc). By this way, it can be concluded

that the hydrogen bonding network between [Bmim][Glu] and (R)-amlodipine is stronger


Acc

epte

d A

rticl

e

than that between [Bmim][Glu] and (S)-amlodipine. The result implies the AAIL is likely to

recognize (R)-amlodipine preferentially. Furthermore, the interaction energies for

[Bmim][Glu] to interact with (R)-amlodipine and (S)-amlodipine are -50.64 and -42.64

kJ/mol, respectively. Considering a more negative energy corresponds to stronger interactions,

the same result can be attained that [Bmim][Glu] interacts more strongly with (R)-amlodipine,

herein its potential recognition effect for (R)-amlodipine is more evident. The above analyses

approve that hydrogen bonding interactions are crucial in determining the enantioselective

recognition capability of AAIL. Similarly, in some discussions earlier, the essential role of

hydrogen bonds to recognize structurally similar bioactive compounds in ionic

liquid-mediated extraction has also been confirmed.24, 43, 44

Optimized geometries of complexes formed by amlodipine enantiomers and other five

AAILs at B3LYP/6-31+G(d,p) level are summarized in Supplementary Figure 3-7. By

comparing these conformations, it can be found that the active sites for amlodipine to interact

with different AAILs are uniform, as has been revealed in the above discussion about

[Bmim][Glu]. It represents that amlodipine has analogous interaction mechanism with these

AAILs. Moreover, the cations interact with the enantiomers mainly through the side carbon

chains, which are also responsible for steric effects. However, the active sites for different

amino-acid derived anions to interact with amlodipine are somewhat different from each

other, due to the subtle change in their compositions and structures.

Interaction energies between different AAILs and amlodipine enantiomers at

B3LYP/6-31+G(d,p) level are listed in Table 1, where ΔER and ΔES refer to energies for

AAILs to interact with (R)-amlodipine and (S)-amlodipine respectively, and the quantity ΔE


Acc

epte

d A

rticl

e

calculated by subtracting ΔER from ΔES represents the energy difference between ΔER and

ΔES. As is shown, ΔER and ΔES for different AAILs are all negative, and vary from -42.35 to

-63.23 kJ/mol. Except for [Bmim][Glu], the values of ΔER for other three AAILs including

[Bmim][Ser], [Emim][Ser] and [Bmim][Phe] are also more negative than the corresponding

ΔES. The result indicates that these AAILs all interact more strongly with (R)-amlodipine

than with (S)-amlodipine, so it is theoretically possible that they all exhibit preferable

recognition effect towards (R)-amlodipine. Unlike the above four AAILs, [Emim][Glu] and

[Emim][Phe] are expected to recognize (S)-amlodipine preferably, because the values of ΔER

for these two AAILs are less negative than ΔES, meaning stronger interactions with

(S)-amlodipine. By further comparing ΔE for all the AAILs, it can be seen this value for

[Bmim][Glu] is very remarkable. It accounts for nearly 20% of the total energy for this AAIL

to interact with (S)-amlodipine, and its absolute value is at least two times that of

[Bmim][Ser], [Emim][Ser], [Bmim][Phe] and [Emim][Phe]. It implies the recognition

capability of [Bmim][Glu] should be much stronger than the other AAILs, because a larger

energy difference between ΔER and ΔES represents a more significant discrepancy between

the stabilities of (S)-amlodipine-AAIL and (R)-amlodipine-AAIL complexes, which is more

beneficial to the separation.

We also calculated the interaction energies based on amlodipine-AAIL complexes

(Supplementary Figure 8-13) optimized at B3LYP-D3/6-311+G(d,p) level, and the results are

summarized in Table 1. As is shown, the values of ΔER and ΔES on this level are significantly

lower than those on B3LYP/6-31+G(d,p) level by 63.87-99.08 and 64.51-96.99 kJ/mol,

respectively, representing that the incorporation of Grimme’s dispersion correction (D3) in


Acc

epte

d A

rticl

e

the DFT based exchange-correlation functions can fairly well consider the dispersion

effects.45, 46

The values of ΔE for [Bmim][Glu], [Bmim][Ser] and [Emim][Ser] are positive,

and that for [Emim][Glu] stay negative, just the same as that observed on B3LYP/6-31+G(d,p)

level, again suggesting their preferable recognition effect for (R)-amlodipine and

(S)-amlodipine, respectively. Most importantly, the absolute value of ΔE for [Bmim][Glu]

nearly doubled compared to that calculated on B3LYP/6-31+G(d,p) level, and is more

significantly higher than for other AAILs, highlighting its potential superior enantioselective

recognition capability for amlodipine racemate.

Table 1. BSSE and ZPE corrected interaction energies calculated on B3LYP/6-31+G(d,p) and

B3LYP-D3/6-311+G(d,p) levels of theory (unit: kJ/mol)

Validation of quantum chemistry calculations

As is mentioned earlier, quantum chemistry calculation has been successfully used to

predict the phase behavior of phenolic homologs in CIL-based extraction systems.24

However,

as far as we know, there is still no report about the predictive capability of quantum chemistry

calculations on the separation of chiral compounds using CILs. Therefore, it is necessary to

validate whether the computational approach would be fit for the enantioselective systems

with experimental data.

On this account, a series of biphasic systems comprising different AAILs were

constructed to experimentally test the chiral recognition capabilities of these chiral

extractants. The resulting distribution coefficients and corresponding enantioselectivities are

displayed in Figure 4a. It can be seen, when [Bmim][Glu], [Bmim][Ser] and [Emim][Ser] are


Acc

epte

d A

rticl

e

used as chiral extractants, the distribution coefficients of (S)-amlodipine are visibly larger

than (R)-amlodipine, and the enantioselectivities are measured as 1.35, 1.11, and 1.05,

respectively. This phenomenon confirms that the three AAILs all interact more strongly with

(R)-amlodipine than with (S)-amlodipine, herein their preferential recognition effects towards

(R)-amlodipine predicted by quantum chemistry calculation are identified. When [Emim][Glu]

is employed, the distribution coefficient of (R)-amlodipine turns out to be higher than that of

(S)-amlodipine, and enantioselectivity below 1 is observed, because of its unique preferable

recognition effect for (S)-amlodipine. The phenomenon that [Emim][Glu] and [Bmim][Glu]

exhibit opposite recognition effect to each other is mainly attributed to the difference in their

lengths of carbon chains in the cations, which greatly influences the steric effect between the

ion pairs and the enantiomers, thus leading to distinct amlodipine-AAIL complex

conformations (Figure 1, Supplementary Figure 3). In the other two Phe--based AAILs

containing systems, the distribution coefficients of (R)-amlodipine are very close to those of

(S)-amlodipine, revealing these AAILs exhibit very poor chiral recognition abilities for

amlodipine. The reason is that both [Bmim][Phe] and [Emim][Phe] have quite close binding

capabilities for different amlodipine enantiomers, with absolute value of ΔE being only 1.75

and 2.02 kJ/mol on B3LYP/6-31+G(d,p), 1.39 and 1.08 kJ/mol on B3LYP-D3/6-311+G(d,p),

as shown in Table 1. Despite there is no meaning in determining the recognition tendency of

[Bmim][Phe] and [Emim][Phe], the result still approves that a larger interaction energy

difference is more favorable for the separation process.

In summary, enantioselectivities of these AAILs investigated in this work are in the order

that [Bmim][Glu] > [Bmim][Ser] > [Emim][Ser] > [Bmim][Phe] ≥ [Emim][Phe] >


Acc

epte

d A

rticl

e

[Emim][Glu]. It is noteworthy that when the enantioselectivity is plotted versus ΔE (Figure

4b), an approximate linear relationship is shown between the two quantities, both on

B3LYP/6-31+G(d,p) and B3LYP-D3/6-311+G(d,p) levels of theory. The result proves that

the pre-screening of AAILs by quantum chemistry calculations based on interaction energies

is applicable to the present work.

Extraction of amlodipine with and without [Bmim][Glu] as chiral extractant

From the above theoretical calculation and experimental results, it has been verified that

[Bmim][Glu] is a promising chiral extractant for amlodipine. In the subsequent studies efforts

would be focused on developing an efficient enantioselective system using [Bmim][Glu] as

the chiral extractant. For this purpose, a collection of organic solvents were combined with

the [Bmim][Glu]-buffer solution to construct different biphasic systems, and phase behaviors

of amlodipine enantiomers in these systems were investigated.

Table 2. Distribution coefficient and enantioselectivity for different biphasic systemsa

From the results summarized in Table 2, it can be seen the distribution coefficients of

amlodipine enantiomers in halohydrocarbon (1,2-dichloroethane/dichloromethane) containing

systems are relatively smaller than in other systems, with or without the addition of

[Bmim][Glu], indicating the affinities of 1,2-dichloroethane and dichloromethane for

amlodipine are somewhat weaker than aliphatic alcohols. In aliphatic alcohols containing

systems, the distribution coefficients of amlodipine enantiomers are in the order that

n-hexanol > n-octanol > n-decanol, which is consistent with the polarities of these aliphatic


Acc

epte

d A

rticl

e

alcohols. When [Bmim][Glu] is absent, the distribution coefficients of (R)-amlodipine and

(S)-amlodipine in n-hexanol-buffer reach up to 75.82 and 79.61. While in n-octanol-buffer,

the distribution coefficients for (R)-amlodipine and (S)-amlodipine decrease dramatically to

36.33 and 37.91 respectively, and those in n-decanol-buffer are even lower. When

[Bmim][Glu] is added, the trend still holds as shown in Table 2. These results reveal that the

property of organic solvent has a great effect on the phase behavior of amlodipine, and

aliphatic alcohols of higher polarities would have stronger affinities for amlodipine

enantiomers. More importantly, it can also be seen that the addition of [Bmim][Glu] lowers

the distribution coefficients no matter in aliphatic alcohol or in halohydrocarbon containing

systems, implying there are strong intermolecular interactions between [Bmim][Glu] and

amlodipine enantiomers, which drive the enantiomers into the aqueous phase.

In n-decanol and n-octanol containing systems, the addition of [Bmim][Glu] is observed

to promote the enantioselectivities to different degrees. When [Bmim][Glu] is absent, the

enantioselectivities in n-decanol-buffer and n-octanol-buffer are only 0.97 and 1.04. While

after the addition of [Bmim][Glu], they increase to 1.35 and 1.21 respectively, indicating

[Bmim][Glu] dose have a prominent chiral recognition effect for (R)-amlodipine as has been

predicted by quantum chemistry calculations. By contrast, in other organic solvents

containing systems, the addition of [Bmim][Glu] does not make very distinct difference in

enantioselectivity. The reason might be that the affinities of halohydrocarbons and n-hexanol

towards amlodipine enantiomers are either too small or too large to achieve a moderate

distribution of the enantiomers between the aqueous phase and the organic phase. In

halohydrocarbons containing systems, the smaller distribution coefficients mean the


Acc

epte

d A

rticl

e

concentration of amlodipine in the aqueous phase was higher than in other systems, so the

probability that [Bmim][Glu] was occupied by competing (S)-amlodipine increased. As for in

n-hexanol containing system, the relatively larger distribution coefficients declare the

concentration of amlodipine in the aqueous phase was lower than other two aliphatic alcohol

containing systems, herein the sufficient interaction between (R)-amlodipine and [Bmim][Glu]

was blocked. The results stress the importance of choosing the modest organic solvent for

extraction separation process.

Influence of solution pH on the enantioselective extraction of amlodipine

For this part, n-decanol-buffer/[Bmim][Glu] system in which the aqueous solutions were

weakly acid with pH values ranging from 3.5 to 6.0 was used to investigate the effect of

solution pH on the enantioselective extraction process. The distribution coefficient and

enantioselectivity were plotted versus the pH of the aqueous solution in Figure 5. Apparently,

the distribution coefficient shows a strong dependence on acidity. The increase in solution pH

results in a continuous growth in distribution coefficients for both enantiomers. Amlodipine is

known as a weak alkaline compound and its pKa is equal to 8.6.47

In aqueous solution, it

could exist in either a molecular form or a cationic form depending on the deprotonation and

protonation behavior of the primary amino group. According to the following

Henderson-Hassaelbach equation,48

][/][log BHAcidBBasepKapH (4)

when the pH value is improved, the percentage of monomers in molecular form increases

gradually, implying more amlodipine monomers are ready to dissolve in the organic phase.

For this reason, the variation in distribution coefficients of amlodipine is estimated to be


Acc

epte

d A

rticl

e

reasonable.

The buffer solution pH also greatly affects the enantioselectivity. When the solution pH

is in the range of 3.5 to 5.5, the enantioselectivity grows continuously and reaches its

maximum at pH=5.5. However, further increase in solution pH brings about a severe decrease

in enantioselectivity, and it drops to only 1.06 when the pH is 6.0. Actually, along with the

rapid increase in distribution coefficient, the amount of amlodipine in the aqueous phase

declines very quickly, and the concentration of amlodipine at pH=6.0 equals only nearly half

of that at pH=5.5. The extra low concentration of amlodipine under increased pH would

impair the interactions between the enantiomers and AAIL so that the dramatic decrease in

enantioselectivity is observed. From above, it can be concluded that the solution pH has a

great effect on the extraction process by influencing the existence form of amlodipine

monomers, and pH=5.5 would be appropriate for the separation process.

Influence of the concentration of AAIL on the enantioselective extraction of amlodipine

To investigate the influence of the concentration of AAIL on the extraction efficiency,

aqueous solutions containing [Bmim][Glu] with different AAIL concentrations varying from

0.01 to 0.20 mol/L were prepared and applied to the extraction process, and the results are

summarized in Figure 6. Along with the increase in the concentration of [Bmim][Glu], the

distribution coefficients for both enantiomers decrease because a higher concentration of

[Bmim][Glu] leads to stronger interactions with amlodipine, and consequently higher

concentration of amlodipine enantiomers in the aqueous phase. It is also found the

enantioselectivity increases under low [Bmim][Glu] concentration and reaches the peak at


Acc

epte

d A

rticl

e

cAAIL=0.025 mol/L. However, further increase in the concentration of [Bmim][Glu] would

reduce the enantioselectivity, because non-preferential recognition for (S)-amlodipine is

included.

Influence of initial concentration of racemic amlodipine on the enantioselective

extraction of amlodipine

The effect of initial concentration of racemic amlodipine on the extraction process is

shown in Figure 7, where the initial concentration was in the range of 0.5 to 3.0 g/L. It can be

observed, both the distribution coefficient and the enantioselectivity rise firstly and then

decline with the increase in the initial concentration of amlodipine. The distribution

coefficients come to the maximum values when camlodipine=2.25 g/L, while the

enantioselectivity comes to its maximum at camlodipine=2.0 g/L. The result verifies that the

initial concentration of amlodipine has an important effect on the phase behavior, and 2.0 g/L

would be a suitable initial concentration for amlodipine.

Influence of temperature on the enantioselective extraction of amlodipine

In general, temperature is an important factor influencing the separation efficiency of

chiral extraction process. Lower temperature is usually more beneficial to improve the

enantioselectivity because interactions between chiral extractants and target molecules

usually get weakened under higher temperature, and the chiral recognition effect will be

impaired as a result.40, 49

For this reason, the previously proposed methods for the extraction

of amlodipine using conventional extractants such as cyclodextrin and tartaric acid


Acc

epte

d A

rticl

e

derivatives all require to be handled at frigid temperature,10, 11

which brings about intensive

energy consumption. As far as we know, enantioselectivities of those methods near room

temperature (298.15 K) would not be higher than 1.14.

In this work, the effect of extraction temperature on the phase behavior of amlodipine

enantiomers in n-decanol-buffer/[Bmim][Glu] was studied with the window opened from

288.15 to 318.15 K. According to the results listed in Figure 8, enantioselectivity reaching up

to 1.38 can be achieved at 288.15 K, whereas it drops only slightly under room temperature

and maintains at 1.35 although higher temperature is unfavorable for separation, owing to the

superior chiral recognition capability of [Bmim][Glu]. By contrast to other extraction systems

comprising conventional chiral extractants, the performance has been significantly improved

by the use of AAIL. More importantly, the desirable separation efficiency under room

temperature proves the present system is conducive to energy saving and consumption

reducing.

Recycling of [Bmim][Glu]

The recycling of ionic liquids is important for developing an economic and

environmentally benign separation process. In this work, the recycling of [Bmim][Glu] was

achieved by regenerating the aqueous phase through backward extraction, with no need to

further purify the AAIL. Note that n-decanol was employed as the organic solvent in

backward extraction process mainly for two reasons. Firstly, n-decanol itself is an adequate

organic solvent for the enantioselective extraction of amlodipine racemate. To keep the

organic solvent consistent in both extraction and backward extraction process is beneficial to

establish easy-handling and continuous operations. Secondly, the backward extraction


Acc

epte

d A

rticl

e

efficiency is very desirable. The result of HPLC analysis shows that after the backward

extraction, the concentration of amlodipine in the aqueous phase is below 0.01 g/L, which is

negligible to the initial concentration of amlodipine at 2.0 g/L.

The performance of the reused [Bmim][Glu] containing aqueous phase within 5 cycles is

shown in Figure 9. Compared to extraction process using fresh aqueous phase, the

enantioselectivity obtained with the regenerated aqueous phase declined a little bit. A

reasonable explanation to the phenomenon is that the regenerated aqueous phase is saturated

with n-decanol, which may subtly influence the phase equilibrium. It was also confirmed by

an additional controlled extraction experiment in which a n-decanol pre-saturated

[Bmim][Glu] containing aqueous solution was used. The results indicate that

enantioselectivity of amlodipine also descends to about 1.24, just similar to that in

regenerated CIL-based system. Nevertheless, considering enantioselectivity in the

regenerated system still stays above 1.20 within 5 cycles, and the distribution coefficients in

different cycles only change very slightly, it can be deduced that the reused [Bmim][Glu] still

remains good separation efficiency for amlodipine.

Conclusions

In this work, a novel enantioselective liquid-liquid extraction system comprised of

n-decanol, NaAc/HAc buffer solution and AAIL was proposed for the separation of

amlodipine. Quantum chemistry calculations were carried out to give a pre-screening to the

candidate AAILs, and revealed that [Bmim][Glu] was supposed to exhibit more preferable

chiral recognition effect for amlodipine enantiomers than other studied AAILs. The


Acc

epte

d A

rticl

e

theoretical calculations also provide comprehensive insights into the chiral recognition

mechanisms of AAILs, and show that hydrogen bonding interactions are essential to the

recognition capabilities of AAILs. To validate the theoretical calculations, phase behaviors of

amlodipine enantiomers in diverse biphasic systems containing different AAILs were

investigated, and we found that the enantioselectivity showed a strong dependence on ΔE. It

demonstrates the quantum chemistry calculations are capable to provide reasonable guidance

to the screening and design of CILs. Subsequently, the impacts of several important factors

on the separation efficiency of amlodipine in the present system were investigated. The

results show that under optimal conditions when the initial concentration of [Bmim][Glu] and

racemic amlodipine are 0.025 mol/L and 2.0 g/L respectively, the solution pH is 5.5 and the

extraction temperature is 288.15 K, enantioselectivity reaching up to 1.38 can be achieved.

When operated at 298.15 K, only a small drop in separation efficiency was observed,

revealing the present system made less strict demand on temperature than conventional

extraction systems containing organic chiral extractants, herein the method was proved to be

more energy-efficient. Lastly, the recycling of [Bmim][Glu] by regenerating the AAIL

containing aqueous phase through backward extraction was studied, and the experimental

data showed that the reused [Bmim][Glu] still remained good separation efficiency. Thus, we

have successfully developed a new potential chiral separation process for amlodipine.

Acknowledgment

The authors are grateful for the financial support from the National Science Foundation of

China (Grant No. 21576231) to this work. The authors gratefully acknowledge the support of


Acc

epte

d A

rticl

e

the Research Computing Center in College of Chemical and Biological Engineering at

Zhejiang University for assistance with the calculations carried out in this work.


Acc

epte

d A

rticl

e

Literature Cited

1. Meredith PA, Elliott HL. Clinical pharmacokinetics of amlodipine. Clin Pharmacokinet.

1992;22:22-31.

2. Sun Y, Fang L, Zhu M, Li W, Meng P, Li L, He Z. A drug-in-adhesive transdermal patch

for S-amlodipine free base: in vitro and in vivo characterization. Int J Pharm.

2009;382:165-171.

3. Zhang X, Loke KE, Mital S, Chahwala S, Hintze TH. Paradoxical release of nitric oxide by

an L-type calcium channel antagonist, the R+ enantiomer of amlodipine. J Cardiovasc

Pharmacol. 2002;39:208-214.

4. Fumio T. Enantiomer Separation: Fundamentals and Practical Methods. New York:

Springer-Verlag, Inc., 2004.

5. Peng Y, Gong T, Zhang K, Lin X, Liu Y, Jiang J, Cui Y. Engineering chiral porous

metal-organic frameworks for enantioselective adsorption and separation. Nat Commun.

2014;5:4406.

6. Gong Z, Duan L, Tang A. Amino-functionalized silica nanoparticles for improved

enantiomeric separation in capillary electrophoresis using carboxymethyl-β-cyclodextrin

(CM-β-CD) as a chiral selector. Microchim Acta. 2015;182:1297-1304.

7. Xie J, Tan Q, Yang L, Lai S, Tang S, Cai C, Chen X. A simple and rapid method for chiral

separation of amlodipine using dual chiral mobile phase additives. Anal Methods.

2014;6:4408-4413.


Acc

epte

d A

rticl

e

8. Gotrane DM, Deshmukh RD, Ranade PV, Sonawane SP, Bhawal BM, Gharpure MM,

Gurjar MK. A novel method for resolution of amlodipine. Org Process Res Dev.

2010;14:640-643.

9. Lee HW, Shin SJ, Yu H, Kang SK, Yoo CL. A novel chiral resolving reagent, bis

((S)-mandelic acid)-3-nitrophthalate, for amlodipine racemate resolution: scalable

synthesis and resolution process. Org Process Res Dev. 2009;13:1382-1386.

10. Sunsandee N, Rashatasakhon P, Ramakul P, Pancharoen U, Nootong K, Leepipatpiboon

N. Selective enantioseparation of racemic amlodipine by biphasic recognition chiral

separation system. Sep Sci Technol. 2014;49:1357-1365.

11. Zhang P, Sun G, Qiu Y, Tang K, Zhou C, Yang C. Equilibrium study on enantioselective

distribution of amlodipine besilate enantiomers in a biphasic recognition chiral extraction

system. J Incl Phenom Macrocycl Chem. 2016;85:127-135.

12. Yang Q, Xing H, Cao Y, Su B, Yang Y, Ren Q. Selective separation of tocopherol

homologues by liquid-liquid extraction using ionic liquids. Ind Eng Chem Res.

2009;48:6417-6422.

13. Schuur B, Verkuijl BJV, Minnaard AJ, de Vries JG, Heeres HJ, Feringa BL. Chiral

separation by enantioselective liquid-liquid extraction. Org Biomol Chem. 2011;9:36-51.

14. Kapnissi-Christodoulou CP, Stavrou IJ, Mavroudi MC. Chiral ionic liquids in

chromatographic and electrophoretic separations. J Chromatogr A. 2014;1363:2-10.

15. Cui X, Yu F, Zhu H, Wu KJ, He CH. Review on Enantioseparation Using Chiral Ionic

Liquids. J Chem Eng Chinese Univ. 2016;30:1229-1240.


Acc

epte

d A

rticl

e

16. Zhang M, Liang X, Jiang S, Qiu H. Preparation and applications of surface-confined

ionic-liquid stationary phases for liquid chromatography. TRAC-Trend Anal Chem.

2014;53:60-72.

17. Tao G, He L, Liu W, Xu L, Xiong W, Wang T, Kou Y. Preparation, characterization and

application of amino acid-based green ionic liquids. Green Chem. 2006;8:639-646.

18. Sun B, Mu X, Qi L. Development of new chiral ligand exchange capillary electrophoresis

system with amino acid ionic liquids ligands and its application in studying the kinetics of

L-amino acid oxidase. Anal Chim Acta. 2014;821:97-102.

19. Meng H, Li S, Xiao L, Li C. Functionalized assembly of solid membranes for chiral

separation using polyelectrolytes and chiral ionic liquid. AIChE J. 2013;59:4772-4779.

20. Kimaru IW, Morris L, Vassiliou J, Savage N. Synthesis and evaluation of L-phenylalanine

ester-based chiral ionic liquids for GC stationary phase ability. J Mol Liq.

2017;237:193-200.

21. Tang F, Zhang Q, Ren D, Nie Z, Liu Q, Yao S. Functional amino acid ionic liquids as

solvent and selector in chiral extraction. J Chromatogr A. 2010;1217:4669-4674.

22. Wu H, Yao S, Qian G, Yao T, Song H. A resolution approach of racemic phenylalanine

with aqueous two-phase systems of chiral tropine ionic liquids. J Chromatogr A.

2015;1418:150-157.

23. Wu D, Zhou Y, Cai P, Shen S, Pan Y. Specific cooperative effect for the enantiomeric

separation of amino acids using aqueous two-phase systems with task-specific ionic

liquids. J Chromatogr A. 2015;1395:65-72.


Acc

epte

d A

rticl

e

24. Yang Q, Xing H, Su B, Bao Z, Wang J, Yang Y, Ren Q. The essential role of

hydrogen-bonding interaction in the extractive separation of phenolic compounds by ionic

liquid. AIChE J. 2013;59:1657-1667.

25. Zhang X, Liu Z, Wang W. Screening of ionic liquids to capture CO2 by COSMO-RS and

experiments. AIChE J. 2010;54:2717-2728.

26. Xing H, Zhao X, Li RL, Yang Q, Su B, Bao Z, Yang Y, Ren Q. Improved efficiency of

ethylene/ethane separation using a symmetrical dual nitrile-functionalized ionic liquid.

ACS Sustainable Chem Eng. 2013;1:1357-1363.

27. Wang S, Stubbs JM, Siepmann JI, Sandler SI. Effects of conformational distributions on

sigma profiles in COSMO theories. J Phys Chem A. 2005;109:11285-11294.

28. Bhattacharyya S, Shah FU. Ether functionalized choline tethered amino acid ionic liquids

for enhanced CO2 capture. ACS Sustainable Chem Eng. 2016;4:5441-5449.

29. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Scalmani

G, Barone V, Petersson GA, Nakatsuji H, Li X, Caricato M, Marenich AV, Bloino J,

Janesko BG, Gomperts R, Mennucci B, Hratchian HP, Ortiz JV, Izmaylov AF, Sonnenberg

JL, Williams-Young D, Ding F, Lipparini F, Egidi F, Goings J, Peng B, Petrone A,

Henderson T, Ranasinghe D, Zakrzewski VG, Gao J, Rega N, Zheng G, Liang W, Hada M,

Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O,

Nakai H, Vreven T, Throssell K, Montgomery JA, Jr., Peralta JE, Ogliaro F, Bearpark MJ,

Heyd JJ, Brothers EN, Kudin KN, Staroverov VN, Keith TA, Kobayashi R, Normand J,

Raghavachari K, Rendell AP, Burant JC, Iyengar SS, Tomasi J, Cossi M, Millam JM,


Acc

epte

d A

rticl

e

Klene M, Adamo C, Cammi R, Ochterski JW, Martin RL, Morokuma K, Farkas O,

Foresman JB, Fox DJ. Gaussian 16, Revision A.03. Wallingford, CT: Gaussian, Inc., 2016.

30. Becke AD. Density-functional thermochemistry. III. The role of exact exchange. J Chem

Phys. 1993;98:5648-5652.

31. Lee C, Yang W, Parr RG. Development of the Colle-Salvetti correlation-energy formula

into a functional of the electron density. Phys Rev B: Condens Matter Mater Phys.

1988;37:785-789.

32. Grimme S, Antony J, Ehrlich S, Krieg H. A consistent and accurate ab initio

parametrization of density functional dispersion correction (DFT-D) for the 94 elements

H-Pu. J Chem Phys. 2010;132:154104.

33. Fileti EE, Chaban VV. The scaled-charge additive force field for amino acid based ionic

liquids. Chem Phys Lett. 2014;616:205-211.

34. Fileti EE, Chaban VV. The force field for imidazolium-based ionic liquids: Novel anions

with polar residues. Chem Phys Lett. 2015;633:132-138.

35. Latosińska JN, Latosińska M, Kasprzak J. 35

Cl-NQR and DFT study of electronic

structure of amlodipine and felodipine vascular-selective drugs from the dihydropyridine

Ca++

antagonists group. Chem Phys Lett. 2008;462:295-299.

36. Zhang Y, Chen X, Wang H, Diao K, Chen J. DFT study on the structure and cation-anion

interaction of amino acid ionic liquid of [C3mim]+[Glu]

−. Journal of Molecular Structure:

THEOCHEM. 2010;952:16-24.

37. Boys SF, Bernardi F. The calculation of small molecular interactions by the differences of

separate total energies. Some procedures with reduced errors. Mol Phys. 1970;19:553-566.


Acc

epte

d A

rticl

e

38. Keith TA. AIMAll, Version 16.10.31. Overland Park, KS, USA: TK Gristmill Software,

2016.

39. Holbach A, Godde J, Mahendrarajah R, Kockmann N. Enantioseparation of chiral

aromatic acids in process intensified liquid-liquid extraction columns. AIChE J.

2015;61:266-276.

40. Tang K, Song L, Liu Y, Pan Y, Jiang X. Separation of flurbiprofen enantiomers by

biphasic recognition chiral extraction. Chem Eng J. 2010;158:411-417.

41. Zhang WY, Cui X, Jiang SX, Song HJ, Zhou RF, Ye XQ, He CH. Enantiseparation of

Ofloxacin Enantimers via Diethyl L-Tartrate Extraction. J Chem Eng Chinese Univ.

2017;31:769-775.

42. Parr RG. In Horizons of Quantum Chemistry. New York: Springer-Verlag, Inc., 1980.

43. Cao Y, Xing H, Yang Q, Su B, Bao Z, Zhang R, Yang Y, Ren Q. High performance

separation of sparingly aqua-/lipo-soluble bioactive compounds with an ionic liquid-based

biphasic system. Green Chem. 2012;14:2617-2625.

44. Dong K, Cao Y, Yang Q, Zhang S, Xing H, Ren Q. Role of hydrogen bonds in

ionic-liquid-mediated extraction of natural bioactive homologues. Ind Eng Chem Res.

2012;51:5299-5308.

45. Zhang Y, He H, Dong K, Fan M, Zhang S. A DFT study on lignin dissolution in

imidazolium-based ionic liquids. RSC Adv. 2017;7:12670-12681.

46. Zhao Y, Truhlar DG. Density functionals with broad applicability in chemistry. Acc Chem

Res. 2008;39:157-167.


Acc

epte

d A

rticl

e

47. Sesin J, Tamargo J. Clinical pharmacokinetics of calcium antagonists. Medicina.

1996;57:451-462.

48. Po HN, Senozan NM. The Henderson-Hasselbalch equation: its history and limitations. J

Chem Educ. 2001;78:1499-1503.

49. Tang K, Zhang P, Pan C, Li H. Equilibrium studies on enantioselective extraction of

oxybutynin enantiomers by hydrophilic β-cyclodextrin derivatives. AIChE J.

2011;57:3027-3036.


Acc

epte

d A

rticl

e

Figure 1. The structures of (a) (R)-amlodipine and (b) (S)-amlodipine

Figure 2. Schematic representation of the experimental procedure


Acc

epte

d A

rticl

e

Figure 3. Optimized structures of (a) (R)-amlodipine-[Bmim][Glu] and (b)

(S)-amlodipine-[Bmim][Glu] at B3LYP/6-31+G(d,p) level. All interatomic distances

represented by dash lines are in angstroms.

Figure 4. (a) Distribution coefficient and enantioselectivity of amlodipine enantiomers in

different n-decanol-buffer/AAIL systems and (b) enantioselectivity plotted versus ΔE

calculated on B3LYP/6-31+G(d,p) and B3LYP-D3/6-311+G(d,p) levels of theory. (The initial

concentration of racemic amlodipine and AAIL were 2.0 g/L and 0.025 mol/L, respectively.

The buffer pH was 5.5. Extraction temperature was 298.15 K. R2 for linear fit of B3LYP and

B3LYP-D3 in (b) are 0.85 and 0.91, respectively.)


Acc

epte

d A

rticl

e

Figure 5. Influence of solution pH on distribution coefficient and enantioselectivity. (The

initial concentration of amlodipine and [Bmim][Glu] were 2.0 g/L and 0.025 mol/L,

respectively. The temperature was 298.15 K.)

Figure 6. Influence of concentration of AAIL on distribution coefficient and

enantioselectivity. (The initial concentration of amlodipine was 2.0 g/L. The solution pH was

5.5. The temperature was 298.15 K.)


Acc

epte

d A

rticl

e

Figure 7. Influence of initial concentration of amlodipine on distribution coefficient and

enantioselectivity. (The concentration of [Bmim][Glu] was 0.025 mol/L. The solution pH was

5.5. The temperature was 298.15 K.)

Figure 8. Influence of extraction temperature on the distribution coefficient and

enantioselectivity. (The initial concentration of racemic amlodipine and [Bmim][Glu] were

2.0 g/L and 0.025 mol/L, respectively. The buffer pH was 5.5.)


Acc

epte

d A

rticl

e

Figure 9. Distribution coefficient and enantioselectivity of amlodipine in different recycle

times. (The initial concentration of racemic amlodipine and [Bmim][Glu] were 2.0 g/L and

0.025 mol/L, respectively. The buffer pH was 5.5. The temperature was 298.15 K.)


Acc

epte

d A

rticl

e

Table 1. BSSE and ZPE Corrected Interaction Energies Calculated on B3LYP/6-31+G(d,p)

and B3LYP-D3/6-311+G(d,p) Levels of Theory (unit: kJ/mol)

AAILs

B3LYP/6-31+G(d,p) B3LYP-D3/6-311+G(d,p)

ΔER ΔES ΔE ΔER ΔES ΔE

[Bmim][Glu] -50.64 -42.64 8.00 -124.20 -108.57 15.63

[Bmim][Ser] -45.99 -42.25 3.74 -123.62 -115.24 8.38

[Emim][Ser] -46.57 -44.02 2.55 -131.94 -129.47 2.47

[Bmim][Phe] -50.30 -48.55 1.75 -144.15 -145.54 -1.39

[Emim][Phe] -42.35 -44.37 -2.02 -141.43 -140.35 1.08

[Emim][Glu] -58.73 -63.23 -4.50 -122.60 -127.74 -5.14


Acc

epte

d A

rticl

e

Table 2. Distribution Coefficient and Enantioselectivity for different Biphasic Systemsa

Biphasic systems DR DS α

1,2-dichloroethane-buffer/[Bmim][Glu] 2.90 2.91 1.00

1,2-dichloroethane-buffer 5.33 5.33 1.00

dichloromethane-buffer/[Bmim][Glu] 4.19 4.21 1.00

dichloromethane-buffer 8.01 8.08 1.01

n-decanol-buffer/[Bmim][Glu] 13.13 17.73 1.35

n-decanol-buffer 23.15 22.46 0.97

n-octanol-buffer/[Bmim][Glu] 19.97 24.18 1.21

n-octanol-buffer 36.33 37.91 1.04

n-hexanol-buffer/[Bmim][Glu] 55.43 54.91 0.99

n-hexanol-buffer 75.82 79.61 1.05

a The initial concentration of racemic amlodipine and chiral extractant were 2.0 g/L and 0.025

mol/L, respectively. The buffer pH was 5.5. The temperature was 298.15 K.


Acc

epte

d A

rticl

e

Documents

Quantum Chemistry Calculation Aided Design of Chiral Ionic